Advanced Glycation End-Products Induce Apoptosis of Vascular Smooth Muscle Cells: A Mechanism for Vascular Calcification

Vascular calcification, especially medial artery calcification, is associated with cardiovascular death in patients with diabetes mellitus and chronic kidney disease (CKD). To determine the underlying mechanism of vascular calcification, we have demonstrated in our previous report that advanced glycation end-products (AGEs) stimulated calcium deposition in vascular smooth muscle cells (VSMCs) through excessive oxidative stress and phenotypic transition into osteoblastic cells. Since AGEs can induce apoptosis, in this study we investigated its role on VSMC apoptosis, focusing mainly on the underlying mechanisms. A rat VSMC line (A7r5) was cultured, and treated with glycolaldehyde-derived AGE-bovine serum albumin (AGE3-BSA). Apoptotic cells were identified by Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining. To quantify apoptosis, an enzyme-linked immunosorbent assay (ELISA) for histone-complexed DNA fragments was employed. Real-time PCR was performed to determine the mRNA levels. Treatment of A7r5 cells with AGE3-BSA from 100 µg/mL concentration markedly increased apoptosis, which was suppressed by Nox inhibitors. AGE3-BSA significantly increased the mRNA expression of NAD(P)H oxidase components including Nox4 and p22phox, and these findings were confirmed by protein levels using immunofluorescence. Dihydroethidisum assay showed that compared with cBSA, AGE3-BSA increased reactive oxygen species level in A7r5 cells. Furthermore, AGE3-induced apoptosis was significantly inhibited by siRNA-mediated knockdown of Nox4 or p22phox. Double knockdown of Nox4 and p22phox showed a similar inhibitory effect on apoptosis as single gene silencing. Thus, our results demonstrated that NAD(P)H oxidase-derived oxidative stress are involved in AGEs-induced apoptosis of VSMCs. These findings might be important to understand the pathogenesis of vascular calcification in diabetes and CKD.


Introduction
Vascular complication is an important aspect of the pathological course of diabetes mellitus, and affects the disease-related morbidity and mortality. For the development of such complications, hyperglycemia is suggested to play a central role. Hence, a long term intensive control of glycemic status is demonstrated to play a pivotal role in the prevention of the complication [1][2][3]. For example, the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions days, calcium content was measured by the O-cresolphthaleincomplexone, and the results were normalized by total protein content. This method is superior to the quantification of the degree of calcium deposition, and suitable for determining late phase of calcium deposition, whereas microscopic imaging method is suitable for the detection of calciprotein particles, which are participated in an early phase of mineralization. The results showed that calcium deposition was significantly increased by AGE3-BSA (131 vs. 399 for cBSA and AGE3-BSA, respectively; p < 0.001) ( Figure 1).

Figure 1.
Glycolaldehyde-derived advanced glycation end-products-bovine serum albumin (AGE3-BSA) (100 µg/mL) increased calcium deposition in a rat vascular smooth muscle cell line and it was inhibited by caspase inhibitor. After reaching confluency, A7r5 cells were incubated with calcification medium containing control BSA (cBSA) or AGE3-BSA in the presence or absence of general caspase inhibitor Z-VAD-FMK (10 µM) or the control Z-FA-FMK (10 µM) for three days. Then, the calcium deposition was measured as described in the Method Section. To determine statistical significance, the results were analyzed by unpaired t-test, and the statistical significance was denoted as follows, ** p < 0.001.
To examine effects of apoptosis on calcium deposition, A7r5 cells were treated with general caspase inhibitor Z-VAD-FMK (10 µM) or the control Z-FA-FMK (10 µM) for three days. AGE3-BSA -induced calcium deposition was significantly inhibited by the treatment with caspase inhibitor (208 vs. 407 for Z-VAD-FMK and Z-FA-FMK, respectively; p < 0.001) ( Figure 1). This suggests that AGE-induced calcium deposition is mediated by apoptotic cell death in VSMCs. Thus, we investigate AGE-induced apoptosis and the mechanism in A7r5 cells.

AGE3 Induced VSMC Apoptosis through NAD(P)H Oxidase Activity
As AGE3-BSA showed maximum apoptotic effect at 100 µg/mL concentration, in all subsequent experiments, we used this dose to investigate about the underlying mechanism of apoptosis.
To examine further about apoptosis, cultured A7r5 cells were incubated with cBSA or AGE3-BSA (100 µg/mL) for three days. After treatment, analysis of apoptosis by TUNEL assay showed that AGE3-BSA markedly increased TUNEL positive cells ( Figure 3a). Interestingly, pretreatment of cells Glycolaldehyde-derived advanced glycation end-products-bovine serum albumin (AGE3-BSA) (100 µg/mL) increased calcium deposition in a rat vascular smooth muscle cell line and it was inhibited by caspase inhibitor. After reaching confluency, A7r5 cells were incubated with calcification medium containing control BSA (cBSA) or AGE3-BSA in the presence or absence of general caspase inhibitor Z-VAD-FMK (10 µM) or the control Z-FA-FMK (10 µM) for three days. Then, the calcium deposition was measured as described in the Method Section. To determine statistical significance, the results were analyzed by unpaired t-test, and the statistical significance was denoted as follows, ** p < 0.001.
To examine effects of apoptosis on calcium deposition, A7r5 cells were treated with general caspase inhibitor Z-VAD-FMK (10 µM) or the control Z-FA-FMK (10 µM) for three days. AGE3-BSA -induced calcium deposition was significantly inhibited by the treatment with caspase inhibitor (208 vs. 407 for Z-VAD-FMK and Z-FA-FMK, respectively; p < 0.001) ( Figure 1). This suggests that AGE-induced calcium deposition is mediated by apoptotic cell death in VSMCs. Thus, we investigate AGE-induced apoptosis and the mechanism in A7r5 cells.

AGE3 Induced VSMC Apoptosis through NAD(P)H Oxidase Activity
As AGE3-BSA showed maximum apoptotic effect at 100 µg/mL concentration, in all subsequent experiments, we used this dose to investigate about the underlying mechanism of apoptosis. To examine further about apoptosis, cultured A7r5 cells were incubated with cBSA or AGE3-BSA (100 µg/mL) for three days. After treatment, analysis of apoptosis by TUNEL assay showed that AGE3-BSA markedly increased TUNEL positive cells ( Figure 3a). Interestingly, pretreatment of cells with NAD(P)H oxidase inhibitor including GKT137831 (20 µM) or VAS2870 (10 µM), markedly decreased the number of TUNEL positive cells (Figure 3a). Quantification analysis also showed that the percentage of TUNEL positive cells in a total cell culture population was significantly increased by AGE3-BSA treatment (1% vs. 83% for cBSA and AGE3-BSA, respectively; p < 0.001), and such effect of AGE3-BSA was greatly inhibited by NAD(P)H oxidase inhibitors (14% and 2% for GKT137831 and VAS2870, respectively) ( Figure 3b). These findings suggest that AGE3-BSA-induced apoptosis of VSMC was mediated by the activation of NAD(P)H oxidase. with NAD(P)H oxidase inhibitor including GKT137831 (20 µM) or VAS2870 (10 µM), markedly decreased the number of TUNEL positive cells (Figure 3a). Quantification analysis also showed that the percentage of TUNEL positive cells in a total cell culture population was significantly increased by AGE3-BSA treatment (1% vs. 83% for cBSA and AGE3-BSA, respectively; p < 0.001), and such effect of AGE3-BSA was greatly inhibited by NAD(P)H oxidase inhibitors (14% and 2% for GKT137831 and VAS2870, respectively) ( Figure 3b). These findings suggest that AGE3-BSA-induced apoptosis of VSMC was mediated by the activation of NAD(P)H oxidase. were measured using an ELISA-based method, as described in the Method section. Apoptosis was found to be increased by AGE3-BSA after treatment for five days. The results are presented here as averages ± SE of at least three independent experiments. The statistical significance of the results was analyzed by one-way ANOVA followed by LDS post-hoc test. Statistical significance was denoted as follows, ** p < 0.001 vs. cBSA

AGE3-BSA Induced Expression of NAD(P)H Oxidase Components and ROS Generation in VSMCs
To evaluate further about the roles of NAD(P)H oxidase in AGE-induced apoptosis of VSMC, we checked the effects of AGE3-BSA on the mRNA expression of the components of NAD(P)H in A7r5 cells. Three days after incubation with 100 µg/mL of AGE3-BSA or cBSA, total RNA was isolated were measured using an ELISA-based method, as described in the Method section. Apoptosis was found to be increased by AGE3-BSA after treatment for five days. The results are presented here as averages ± SE of at least three independent experiments. The statistical significance of the results was analyzed by one-way ANOVA followed by LDS post-hoc test. Statistical significance was denoted as follows, ** p < 0.001 vs. cBSA. with NAD(P)H oxidase inhibitor including GKT137831 (20 µM) or VAS2870 (10 µM), markedly decreased the number of TUNEL positive cells (Figure 3a). Quantification analysis also showed that the percentage of TUNEL positive cells in a total cell culture population was significantly increased by AGE3-BSA treatment (1% vs. 83% for cBSA and AGE3-BSA, respectively; p < 0.001), and such effect of AGE3-BSA was greatly inhibited by NAD(P)H oxidase inhibitors (14% and 2% for GKT137831 and VAS2870, respectively) ( Figure 3b). These findings suggest that AGE3-BSA-induced apoptosis of VSMC was mediated by the activation of NAD(P)H oxidase. were measured using an ELISA-based method, as described in the Method section. Apoptosis was found to be increased by AGE3-BSA after treatment for five days. The results are presented here as averages ± SE of at least three independent experiments. The statistical significance of the results was analyzed by one-way ANOVA followed by LDS post-hoc test. Statistical significance was denoted as follows, ** p < 0.001 vs. cBSA

AGE3-BSA Induced Expression of NAD(P)H Oxidase Components and ROS Generation in VSMCs
To evaluate further about the roles of NAD(P)H oxidase in AGE-induced apoptosis of VSMC, we checked the effects of AGE3-BSA on the mRNA expression of the components of NAD(P)H in A7r5 cells. Three days after incubation with 100 µg/mL of AGE3-BSA or cBSA, total RNA was isolated A7r5 cells were incubated in calcification medium containing cBSA, or AGE3-BSA (100 µg/mL) in the presence or absence of NAD(P)H oxidase inhibitors including GKT137831 (20 µM) or VAS2870 (10 µM) for three days. Apoptosis was evaluated by TUNEL assay, as described in the Method section. Cells in the culture were identified by nuclear staining with Hoechst, and evaluated under a fluorescence microscope; (b) For quantification, Hoechst and TUNEL double positive cells were counted in 10 random microscopic fields at 200× magnification, and expressed as percent TUNEL positive cells in a culture. Statistical significance of the results was analyzed by one-way ANOVA followed by LDS post-hoc test. Statistical significance was denoted as follows, ** p < 0.001 vs. AGE3-BSA.

AGE3-BSA Induced Expression of NAD(P)H Oxidase Components and ROS Generation in VSMCs
To evaluate further about the roles of NAD(P)H oxidase in AGE-induced apoptosis of VSMC, we checked the effects of AGE3-BSA on the mRNA expression of the components of NAD(P)H in A7r5 cells. Three days after incubation with 100 µg/mL of AGE3-BSA or cBSA, total RNA was isolated from A7r5 cells, and the mRNA expression of Nox1, Nox4 and p22 phox was assessed by real-time PCR. The results showed that AGE3-BSA (100 µg/mL) treatment significantly increased the expression of Nox1, Nox4 and p22 phox mRNA (24%; p < 0.05, 43%; p < 0.05, and 51%; p < 0.05, respectively) ( Figure 4). from A7r5 cells, and the mRNA expression of Nox1, Nox4 and p22 phox was assessed by real-time PCR. The results showed that AGE3-BSA (100 µg/mL) treatment significantly increased the expression of Nox1, Nox4 and p22 phox mRNA (24%; p < 0.05, 43%; p < 0.05, and 51%; p < 0.05, respectively) ( Figure 4). A7r5 cells were treated with cBSA or AGE3-BSA (100 µg/mL) for three days. Total mRNA was isolated, and Nox1, Nox4 and p22 phox mRNA levels were evaluated by real-time PCR, as described in the Method section. GAPDH mRNA was used as a loading control. The data presented here as averages ± SE of at least three experiments. The statistical significance of the results was analyzed by unpaired t-test. The statistical significance was denoted as follows, * p < 0.05 vs. cBSA.
We further analyzed the expression of Nox4 and p22 phox at the protein level. Immunofluorescence results demonstrated that Nox4 protein level was barely detectable in medium treated cells both at Day 3 and 5, whereas p22 phox level was quite high at Day 3, which was decreased at Day 5. AGE3-BSA treatment increased Nox4 protein both at Day 3 and 5 (Figure 5a,b,e). In the case of p22 phox , the protein level was increased slightly, but significantly by AGE3-BSA at Day 3 comparing medium treated and cBSA treated cells (Figure 5c  A7r5 cells were treated with cBSA or AGE3-BSA (100 µg/mL) for three days. Total mRNA was isolated, and Nox1, Nox4 and p22 phox mRNA levels were evaluated by real-time PCR, as described in the Method section. GAPDH mRNA was used as a loading control. The data presented here as averages ± SE of at least three experiments. The statistical significance of the results was analyzed by unpaired t-test. The statistical significance was denoted as follows, * p < 0.05 vs. cBSA.
We further analyzed the expression of Nox4 and p22 phox at the protein level. Immunofluorescence results demonstrated that Nox4 protein level was barely detectable in medium treated cells both at Day 3 and 5, whereas p22 phox level was quite high at Day 3, which was decreased at Day 5. AGE3-BSA treatment increased Nox4 protein both at Day 3 and 5 (Figure 5a,b,e). In the case of p22 phox , the protein level was increased slightly, but significantly by AGE3-BSA at Day 3 comparing medium treated and cBSA treated cells (Figure 5c The results showed that AGE3-BSA (100 µg/mL) treatment significantly increased the expression of Nox1, Nox4 and p22 phox mRNA (24%; p < 0.05, 43%; p < 0.05, and 51%; p < 0.05, respectively) ( Figure 4). We further analyzed the expression of Nox4 and p22 phox at the protein level. Immunofluorescence results demonstrated that Nox4 protein level was barely detectable in medium treated cells both at Day 3 and 5, whereas p22 phox level was quite high at Day 3, which was decreased at Day 5. AGE3-BSA treatment increased Nox4 protein both at Day 3 and 5 (Figure 5a,b,e). In the case of p22 phox , the protein level was increased slightly, but significantly by AGE3-BSA at Day 3 comparing medium treated and cBSA treated cells (Figure 5c,f). However, AGE3-BSA considerably increased p22 phox protein level at Day 5 (Figure 5d,f).  Next, we checked whether such increased production of Nox4 and p22 phox proteins by AGE3-BSA has any functional significance. A7r5 cells were treated with medium, cBSA or AGE3-BSA, and cellular ROS levels were checked by dihydroethidium (DHE) assay. The results showed that AGE3-BSA significantly increased cellular ROS level compared to medium treated and cBSA treated conditions (Figure 5g,h).

Silencing Nox4 and p22 phox Suppressed AGE3-Induced Apoptosis of A7r5 Cells
Next, we examined effects of the silencing of Nox4 and p22 phox on AGE-induced apoptosis using mRNA-specific siRNA transfection in A7r5 cells. The real time PCR results showed that both Nox4 and p22 phox mRNA levels were decreased to 5%-20% after mRNA specific siRNA transfection, indicating their sufficient silencing effect (Figure 6a). Importantly, AGE3-BSA-induced A7r5 apoptosis was markedly inhibited (42% or 47%) by transfection of either Nox4 or p22 phox siRNA, compared to control (scramble siRNA transfection) (Figure 6b). However, we did not find any synergistic or additive effect of double silencing of Nox4 and p22 phox on the apoptosis (42% inhibition; not significant vs. Nox4 siRNA or p22 phox siRNA transfection). Next, we checked whether such increased production of Nox4 and p22 phox proteins by AGE3-BSA has any functional significance. A7r5 cells were treated with medium, cBSA or AGE3-BSA, and cellular ROS levels were checked by dihydroethidium (DHE) assay. The results showed that AGE3-BSA significantly increased cellular ROS level compared to medium treated and cBSA treated conditions (Figure 5g,h).

Silencing Nox4 and p22 phox Suppressed AGE3-Induced Apoptosis of A7r5 Cells
Next, we examined effects of the silencing of Nox4 and p22 phox on AGE-induced apoptosis using mRNA-specific siRNA transfection in A7r5 cells. The real time PCR results showed that both Nox4 and p22 phox mRNA levels were decreased to 5%-20% after mRNA specific siRNA transfection, indicating their sufficient silencing effect (Figure 6a). Importantly, AGE3-BSA-induced A7r5 apoptosis was markedly inhibited (42% or 47%) by transfection of either Nox4 or p22 phox siRNA, compared to control (scramble siRNA transfection) (Figure 6b). However, we did not find any synergistic or additive effect of double silencing of Nox4 and p22 phox on the apoptosis (42% inhibition; not significant vs. Nox4 siRNA or p22 phox siRNA transfection). Figure 6. AGE3-BSA-induced apoptosis is mediated through Nox4 or p22 phox . (a) Nox4 or p22 phox siRNA was transfected to A7r5 cells, total RNA was isolated three days after transfection, and Nox4 and p22 phox mRNA levels were measured by real-time PCR, as described in the Method section. GAPDH mRNA was used as loading control. A scramble siRNA was used as a negative control (NC); (b) AGE3-BSA-induced apoptosis was inhibited by silencing of Nox4 and p22 phox mRNA. Apoptosis was evaluated by cell death detection ELISA kit. The absorbance in cells treated with AGE3-BSA was compared with that in cells treated with cBSA. The ratio was demonstrated after correction with each cBSA. Results were analyzed by one-way ANOVA and LDS post-hoc test. ** p < 0.001 vs. NC.

Discussion
We have previously shown that AGEs stimulate calcium deposition in VSMCs through excessive ROS generation and phenotypic transition to osteoblastic cells [19]. In the present study, we observed that AGE3-BSA significantly induced calcium deposition and apoptosis in A7r5 cells. The mRNA and protein expression of Nox4 and p22 phox was up-regulated by AGE3-BSA treatment. Moreover, AGE3induced apoptosis was significantly suppressed by pretreating the cells with Nox inhibitors. Nox4 and p22 phox silencing showed similar inhibitory effects on the apoptosis. Taken together, AGE3induced apoptosis of VSMCs might be the result of NAD(P)H oxidase activation and ROS generation.
In vasculature, ROS is generated mainly by NAD(P)H oxidase. In a previous study, oxidized LDL causes NAD(P)H oxidase mediated excessive ROS generation, resulting accelerated apoptosis of VSMCs [20]. Diabetic model mice showed elevated expression of Nox, bone morphogenetic protein-4, and osteopontin (OPN) genes in the aorta, which was suppressed by TEMPOL, a superoxide scavenger [21]. Patients with type 1 diabetes showed a positive association between the concentration of pentosidine, one of AGEs, and 8-OHdG in the urine [22]. We have previously observed that 8-OHdG level in the culture medium was elevated by AGE3 treatment in VSMCs, indicating that AGE stimulates ROS production [19]. Since excessive ROS is thought to stimulate the production of AGEs [23,24], the generation of AGEs and ROS might activate a positive feedback loop. NAD(P)H oxidase is composed of Nox isoforms, p22 phox , and associated proteins such as p47 phox as a subunit. According to recent studies, Nox1, Nox4, p22 phox and p47 phox are expressed in VSMCs, and Nox4 and p22 phox possess significant functionality among them [25][26][27][28][29][30]. Using gene targeting mouse model, their functionality in the vessel has been proved; p22 phox is involved in the progression of atheroma [31], while Nox4 facilitates cardiac adaptation to chronic stress such as pressure overload or hypoxia [32]. In addition, Nox4 plays a key role in the pathogenesis of diabetic nephropathy by targeting renal fumarate hydratase, the enzyme that increases fumarate levels [33]. In the present study, AGE3-induced apoptosis was suppressed by silencing Nox4 or p22 phox , and double silencing did not show any synergistic, or even additive effects, indicating that the presence of both Nox4 and p22 phox is essential for functional activity of NAD(P)H oxidase. Moreover, the function of one of them cannot compensate the other components. As previous studies demonstrated that the activity of p22 phox as well as Nox4 is associated with their mRNA levels [34,35], increased expression of Nox4 or p22 phox is most probably responsible for ROS generation. Since silencing of Nox1 did not affect AGEinduced calcium deposition in our previous study [19], we speculate that expression level of Nox1 is of little importance for calcium deposition in VSMCs. Future studies need to address the roles of Nox1 and p47 phox in VSMCs. Figure 6. AGE3-BSA-induced apoptosis is mediated through Nox4 or p22 phox . (a) Nox4 or p22 phox siRNA was transfected to A7r5 cells, total RNA was isolated three days after transfection, and Nox4 and p22 phox mRNA levels were measured by real-time PCR, as described in the Method section. GAPDH mRNA was used as loading control. A scramble siRNA was used as a negative control (NC); (b) AGE3-BSA-induced apoptosis was inhibited by silencing of Nox4 and p22 phox mRNA. Apoptosis was evaluated by cell death detection ELISA kit. The absorbance in cells treated with AGE3-BSA was compared with that in cells treated with cBSA. The ratio was demonstrated after correction with each cBSA. Results were analyzed by one-way ANOVA and LDS post-hoc test. ** p < 0.001 vs. NC.

Discussion
We have previously shown that AGEs stimulate calcium deposition in VSMCs through excessive ROS generation and phenotypic transition to osteoblastic cells [19]. In the present study, we observed that AGE3-BSA significantly induced calcium deposition and apoptosis in A7r5 cells. The mRNA and protein expression of Nox4 and p22 phox was up-regulated by AGE3-BSA treatment. Moreover, AGE3-induced apoptosis was significantly suppressed by pretreating the cells with Nox inhibitors. Nox4 and p22 phox silencing showed similar inhibitory effects on the apoptosis. Taken together, AGE3-induced apoptosis of VSMCs might be the result of NAD(P)H oxidase activation and ROS generation.
In vasculature, ROS is generated mainly by NAD(P)H oxidase. In a previous study, oxidized LDL causes NAD(P)H oxidase mediated excessive ROS generation, resulting accelerated apoptosis of VSMCs [20]. Diabetic model mice showed elevated expression of Nox, bone morphogenetic protein-4, and osteopontin (OPN) genes in the aorta, which was suppressed by TEMPOL, a superoxide scavenger [21]. Patients with type 1 diabetes showed a positive association between the concentration of pentosidine, one of AGEs, and 8-OHdG in the urine [22]. We have previously observed that 8-OHdG level in the culture medium was elevated by AGE3 treatment in VSMCs, indicating that AGE stimulates ROS production [19]. Since excessive ROS is thought to stimulate the production of AGEs [23,24], the generation of AGEs and ROS might activate a positive feedback loop. NAD(P)H oxidase is composed of Nox isoforms, p22 phox , and associated proteins such as p47 phox as a subunit. According to recent studies, Nox1, Nox4, p22 phox and p47 phox are expressed in VSMCs, and Nox4 and p22 phox possess significant functionality among them [25][26][27][28][29][30]. Using gene targeting mouse model, their functionality in the vessel has been proved; p22 phox is involved in the progression of atheroma [31], while Nox4 facilitates cardiac adaptation to chronic stress such as pressure overload or hypoxia [32]. In addition, Nox4 plays a key role in the pathogenesis of diabetic nephropathy by targeting renal fumarate hydratase, the enzyme that increases fumarate levels [33]. In the present study, AGE3-induced apoptosis was suppressed by silencing Nox4 or p22 phox , and double silencing did not show any synergistic, or even additive effects, indicating that the presence of both Nox4 and p22 phox is essential for functional activity of NAD(P)H oxidase. Moreover, the function of one of them cannot compensate the other components. As previous studies demonstrated that the activity of p22 phox as well as Nox4 is associated with their mRNA levels [34,35], increased expression of Nox4 or p22 phox is most probably responsible for ROS generation. Since silencing of Nox1 did not affect AGE-induced calcium deposition in our previous study [19], we speculate that expression level of Nox1 is of little importance for calcium deposition in VSMCs. Future studies need to address the roles of Nox1 and p47 phox in VSMCs.
According to previous works, various mechanisms such as growth arrest specific gene 6 (gas6) mediated Axl-PI3kinase-Akt pathway, osteoprotegerin/RANKL-RANK system, activation of matrix metalloproteinases, inflammatory cytokines, and inhibiting factors including pyrophosphate, matrix Gla protein, and α2-HS-glycoprotein (fetuin-A) are involved in the development of vascular calcification [36][37][38][39]. Regarding signal transduction, Tanikawa and coworkers indicated that AGEs-induced VSMC calcification is mediated through RAGE and p38 MAPK pathway [40]. Since activation of p38 MAPK is associated with excessive ROS production and cellular apoptosis [41], there is a possibility that AGE-induced apoptosis promotes vascular calcification through the RAGE/p38 MAPK pathway. On the other hand, high concentration of extracellular phosphate induced apoptosis and calcium deposition in VSMCs through Gas6-Axl interaction [36]. It was demonstrated that atorvastatin inhibited calcification by preventing apoptosis without affecting mevalonate pathway. Although the molecular mechanisms of apoptosis and vascular calcification may differ between AGEs and phosphate, further study is necessary to elucidate the mechanisms involved in the pathogenesis of AGEs-induced or ROS-mediated calcification of VSMCs.
In medial calcification, loss of VSMCs has been observed. In uremic model mice, VSMC phenotype change and VSMC loss was observed earlier than the progression of calcification [42]. In the aorta of CKD patients, apoptotic cell death was observed in vascular calcified area [43]. In diabetes or hyperglycemic condition, VSMCs exhibited significantly increased rates of proliferation, apoptosis, and migration, and decreased expression of contractile regulating proteins as well as abnormal cellular morphology, suggesting that normal vascular structure and function are impaired [44][45][46][47][48]. Therefore, prevention of AGE generation or AGE-RAGE interaction can be a therapeutic target of the progression of arteriosclerosis including vascular calcification [13]. In addition, a very recent study has shown that AGE-induced VSMC apoptosis and ER stress are augmented by treatment with a novel carboxymethylated peptide [46].

Cell Culture
A7r5 cell, a rat aortic VSMC line, was obtained from European Collection of Cell Cultures through Dainippon Seiyaku (Osaka, Japan). The cells were cultured in fully humidified atmospheric air containing 5% CO 2 condition at 37 • C temperature, with Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS, 100 U/mL penicillin and 100 U/mL streptomycin.

Induction of Calcification
Calcification of A7r5 was induced following the method described by Shioi et al. [49]. Briefly, after reaching confluency, the growth medium of A7r5 was changed to calcification medium (DMEM containing 10% FBS, 10 mM sodium pyruvate, 10 −7 M insulin, 100 U/mL penicillin, 100 mg/mL streptomycin, and β-glycerophosphate). The medium was replaced with fresh medium twice a week. Treatment with general caspase inhibitor Z-VAD-FMK (Abcam, Cambridge, UK) or the control Z-FA-FMK (Abcam) was performed to examine an effect of apoptosis on calcium deposition.

Preparation of AGEs
AGE-BSA was prepared as described previously [50]. Briefly, BSA was incubated under sterile conditions with glycolaldehyde (AGE3) (Sigma Aldrich, St. Louis, MO, USA) and 5 mM diethylenetriamine pentaacetic acid (DTPA) in 0.2 M phosphate buffer (pH 7.4) at 37 • C for seven days. Then, the low molecular weight reactants and aldehydes were removed using a PD-10 chromatography column and dialysis against PBS.

Quantification of Calcium Deposition
Cells were decalcified with 0.6 N HCl for 24 h. The calcium content of HCl supernatant was determined colorimetrically by o-cresolphthaleincomplexone method (calcium C-test Wako; Wako Pure Chemical Industries, Osaka, Japan). After decalcification, cells were washed three times with PBS and solubilized with 0.1 N NaOH/0.1% SDS. Protein content was measured with the Pierce BCA Protein Assay kit (Thermo Fisher Scientific, Waltham, MA, USA). Calcium content of the cell layer was normalized by protein content.

TUNEL Assay
Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) was done using a kit according to the manufacturer's protocol (In Situ Cell Death Detection Kit, POD, Roche Molecular Biochemicals, Mannheim, Germany). Briefly, A7r5 cells were seeded on a chamber slide at a density of 2 × 10 4 cells/well and cultured overnight in DMEM with 10% FBS and antibiotics. On the next day, the cells were treated with AGE3 or control BSA in the presence or absence of NAD(P)H oxidase (Nox) inhibitors, GKT137831 (Selleck Chemicals, Houston, TX, USA) or VAS2870 (Sigma Aldrich) for three days. After fixation with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100, the DNA nicks in apoptotic cells were labeled with fluorescein-conjugated nucleotides using labeling solutions provided by the manufacturer. To identify cells, nuclei were stained with Hoechst. Apoptosis was semi-quantitatively evaluated by the ratio of TUNEL positive cell number divided by Hoechst positive cell number. Briefly, 10 fields were randomly selected in each staining of each slide, and the positive cell number was counted by two different observers under the fluorescence microscope at the same condition to calculate the ratio.

Apoptosis Assay Using a DNA Fragment Detection ELISA Kit
A7r5 cells were seeded on a 96-well plate at a density of 2 × 10 4 cells/well, and cultured overnight in DMEM with 10% FBS and antibiotics. On the next day, the cells were treated with AGE3 or control BSA (cBSA). After five days, the cells were lysed and the supernatant was analyzed for DNA fragments using an enzyme-linked immunosorbent assay (ELISA) kit, according to the manufacturer's protocol (Cell Death Detection ELISA, Roche Molecular Biochemicals).

Quantification of mRNA Expression by Real-Time PCR
Total RNA was isolated from cultured A7r5 cells using Trisol reagent (Invitrogen, San Diego, CA, USA) according to the manufacturer's recommended protocol. First-strand cDNA was synthesized using oligo-dT primer and SuperScript III cDNA synthesis kit (Invitrogen). SYBR green chemistry was used to perform quantitative determination of the mRNAs, following an optimized protocol [51]. The design of sense and antisense oligonucleotide primers was based on published cDNA sequences using the Primer Express software (version 2.0.0, Applied Biosystems, Carlsbad, CA, USA). The cDNA was amplified using an ABI PRISM 7000 sequence detection system (Applied Biosystems). The cDNA-specific SYBR Green Mix was incorporated into the PCR buffer provided in the QuantiTect SYBR PCR kit (QIAGEN, Valencia, CA, USA) to allow for quantitative detection of the PCR product. The temperature profile of the reaction was 60 • C for 2 min, followed by 95 • C for 15 min, and 40 cycles of denaturation at 94 • C for 15 s, and annealing and extension at 60 • C for 1 min.

RNA Interference
RNA interference was used to down-regulate the expression of Nox4 and p22 phox in A7r5 cells. SMARTpool small interfering RNA (siRNA) and SMARTpool reagents for these genes, and nonspecific control siRNA duplexes were designed and synthesized by Customer SMARTpool siRNA Design from Dharmacon (Lafayette, CO, USA). For gene knock down experiments, A7r5 cells were plated in 1.5 cm dish and cultured for 24 h in DMEM containing 10% FBS and antibiotics. Next, after 24 h incubation in medium without antibiotics, cells were transfected with siRNAs (25 nM) using transfection reagent according to the manufacturer's instructions. After another 48 h of culture, cells were incubated in calcification medium.

Immunofluorescence
A7r5 cells were cultured in a Lab-Tek II Chamber Slide (Thermo Fisher Scientific Inc., Yokohama, Japan). After appropriate treatment, the medium was removed, cells were washed once with PBS, and fixed with 100% methanol. After blocking with a solution containing 0.1% Triton-X100, and normal goat of horse serum, the cells were incubated with anti-NOX4 IgG (goat, 1:200, Santa Cruz, Dallas, TX, USA), or anti-p22 phox IgG (rabbit, 1:200, Santa Cruz) for 2 h at room temperature. FITC conjugated anti-goat IgG, and Texas red conjugated anti-rabbit IgG were used to detect Nox4 and p22 phox , respectively. To identify the cells, nuclei were visualized with Hoechst staining. Then stained cells were examined under a fluorescence microscope, and photographs were taken. The fluorescence intensities were quantified using ImageJ software (http://imagej.net/ImageJ).

DHE Measurement
To determine cellular ROS levels, DHE assay was done following a previously described method [52]. Briefly, after treatment, cells were washed once with PBS, and incubated in serum free medium containing 30 µM DHE for 3 min at room temperature on a rocking platform after protecting from light. Then the cells were washed once with PBS and quickly examined under a fluorescence microscope and photographs were taken. The DHE fluorescence intensities were quantified using ImageJ software.

Statistics
The numerical data are expressed as mean ± SEM. Statistical evaluation of the differences between the groups was carried out with unpaired t-test and/or one-way analysis of variance (ANOVA) followed by Fisher's protected least significant difference (LSD). For all statistical tests, a value of p < 0.05 was considered to be statistically significant difference.

Conclusions
AGEs stimulate VSMC apoptosis through excessive ROS generation. Component of NAD(P)H oxidase such as Nox4 may be a good candidate of new strategy to prevent vascular calcification.